The long-term goal of this project is to develop an HIV vaccine or vaccine component using genetically detoxified bacterial toxins as vector molecules for induction of CD8+ T cell responses, including cytolytic T lymphocytes (CTL). The candidate vaccine molecules will consist of genetic fusions of HIV antigens with the toxins, to stimulate HIV-specific CTL activity and other T-cell responses in vivo. Bacterial toxins are suitable vector molecules because they can enter eukaryotic cells and, therefore, deliver antigens for processing and presentation by MHC class I molecules, a complex which is recognized by CTL. This property of the toxins is important because only internalized (cytosolic) antigens are efficiently targeted to the MHC class I pathway. The toxins to be used are genetically detoxified derivatives of pertussis toxin (PT) and choler toxin (CT), PT-9K/129G and CT-K63, respectively, which are completely devoid of toxicity but retain other properties. The fusion molecules will be tested in model systems in vitro and in vivo to assess their MHC class I targeting capacity and their immunogenicity. The specific aims of the proposed research are as follows: (1) to construct detoxified toxin- HIV/SIV antigen fusions that are assembled and secreted efficiently by the bacterial hosts, and to purify these fusion proteins; (2) to demonstrate the CTL-stimulatory capacity of such fusion proteins in vitro, by infection of target antigen presenting cells and CTL lysis assays, and in vivo, by immunization of mice and rhesus macaques, and assay for antigen- specific CTL and other immune responses. one advantage of such strategy is that it utilizes a vector that, in one case, is already a vaccine molecule (PT-9K/129G, which is in use as a pertussis vaccine component), and in the other case (CY), is widely used as an adjuvant, particularly for enhancement of mucosal immune responses. These fusions may, therefore, represent powerful mucosal and systemic vaccine molecules for HIV.